<<That's your algorithm and TI slowness at work.>> Auugh. As I've said REPEATEDLY, Wheel Factorization is better than Baby Trial Divide but worse than some of the more esoteric selections in Number Theory. I'm sure you can't do better yourself.